State-of-the-Art in Nudity Classification: A Comparative Analysis
Akyon, Fatih Cagatay, Temizel, Alptekin
–arXiv.org Artificial Intelligence
Stable This paper presents a comparative analysis of existing nudity Diffusion safety checker is designed to prevent unsafe image classification techniques for classifying images based generation, while LAION safety checker works by filtering on the presence of nudity, with a focus on their application out unwanted images from the training set to prevent diffusion in content moderation. The evaluation focuses on models being trained on inappropriate images. These CNN-based models, vision transformer, and popular opensource safety checkers demonstrate the growing importance of developing safety checkers from Stable Diffusion and Largescale effective and accurate image classification systems Artificial Intelligence Open Network (LAION).
arXiv.org Artificial Intelligence
Dec-26-2023
- Country:
- Asia > Middle East > Republic of Türkiye > Ankara Province > Ankara (0.05)
- Genre:
- Research Report (1.00)
- Industry:
- Leisure & Entertainment (0.47)
- Media > Film (0.47)
- Technology: